1

Configuration and Performance of IEC 61850 for First-Time Users – UNC Charlotte Senior

Design Project Youssef Botza, Matthew Shaw, Peter Allen, Mike Staunton, and Dr. Robert Cox,

University of North Carolina at Charlotte Michael Boughman, Casey Roberts, and William Rominger, Schweitzer Engineering Laboratories, Inc.

Abstract—IEC 61850 was designed to be a substation IED network and system communications standard rather than just another communications protocol. The international standard design allows for the interoperability of many different intelligent electronic devices (IEDs). Using Ethernet, IEDs are networked within a substation and across an entire power network. The University of North Carolina Charlotte senior design group applied the IEC 61850 standard to a substation integration project that was first designed using traditional serial communi-cations methods. The purpose of this project was to offer a practical comparison between a system of protective relays communicating protection schemes with serial communications and hardwired contacts and another using IEC 61850 GOOSE (Generic Object Oriented Substation Event) messages via Ethernet. Because this project was an introduction to power systems for most of the design team, the practical implementation of incorporating IEC 61850 into a substation integration project is presented at a beginner’s level.

I. NOMENCLATURE The following list contains the definitions of abbreviations

that are used in this paper. AX-S4 MMS IEC 61850 MMS Client Interface Software COM Component Object Model DCOM Distributed Component Object Model DDE Dynamic Data Exchange GOOSE Generic Object Oriented Substation Event GUI Graphical User Interface HMI Human-Machine Interface IED Intelligent Electronic Device LAN Local-Area Network MMS Manufacturing Message Specification MOE MMS Object Explorer MTBF Mean Time Between Failures OLE Object Linking and Embedding OPC OLE for Process Control RB Remote Bit SCADA Supervisory Control and Data Acquisition SER Sequential Events Recorder SLC Software Logic Controller TCP/IP Transmission Control Protocol/Internet

Protocol WWMMELink Software Application Script That

Concatenates WWMMEItem1 and WWMMEItem2

II. OVERVIEW OF IEC 61850 IEC 61850 was created to be an internationally standard-

ized method of communication and integration. The standard is intended to allow IEDs from multiple manufacturers to be networked to perform protection, monitoring, automation, metering, and control. IEC 61850 supports all substation automation functions and the engineering required for imple-mentation. Unlike earlier standards, the technical approach of IEC 61850 was premeditated to make it flexible and allow for future improvements.

An effort to create a communications standard with global appeal was initiated by an international IEC project group of about 60 members. In 1995, they created Technical Committee 57 to produce international standards in the field of communications between the equipment and systems for the electric power process, including telecontrol, teleprotec-tion, and all other telecommunications to control the electric power system. This committee began creating the communi-cations standard IEC 61850. The objectives set for the standard were:

• Develop a standard comprised of multiple protocols for complete substation communication.

• Define basic services required to transfer data. • Promote a high level of interoperability between

devices and systems from different manufacturers. • Create a common method and format for modeling,

describing, and exposing data. • Standardize configuration file structure and content to

simplify device configuration and methods for sharing configuration parameters among devices and systems.

The IEC 61850 standard specifies the data transfer methods and the server processes within the substation; this process is based on a hierarchical data structure with an object-oriented approach. The data objects are grouped by functional constraints to allow information to be communicated with high-efficiency data transfer. This standard also promotes interoperability, where multiple IED manufacturers can communicate over one or several standardized protocols.

Throughout the years, there have been many protocols used within substations. These protocols are often proprietary with custom communications links, which can make interopera-bility between multiple manufacturers’ IEDs difficult. Using

2

IEC 61850 simplifies the interoperation of devices from different manufacturers.

IEC 61850 differs from most previous communications methods in its use of object models for device functions and device components. These models define common data formats, identifiers, and controls for substation and feeder devices such as meters, switches, voltage regulators, and protection relays. The models specify standardized and logical groups of data for the most common device functions and allow for significant manufacturer specialization.

III. PROJECT AND SYSTEM OVERVIEW A student team from University of North Carolina

Charlotte conducted a senior design project that made a practical comparison between a system of protection relays using hardwired contacts and serial communications, and one using IEC 61850 GOOSE messages via Ethernet. The project system consisted of ten IEDs networked to provide protection, monitoring, automation, metering, and control of two 138 kV lines, a 138 kV ring bus, a 12.47 kV feeder, and a transformer. The team provided a fully automated system with a rugged computing platform and network switch that provided SCADA and remote engineering control using LAN access. The integration of the project resulted in pertinent IED information displayed on an HMI with a GUI.

The group performed a comparison between IEC 61850 and the traditional methods of wiring microprocessor-based protective relays. By designing and drafting a complete set of drawings for each system, they were able to perform both quantitative and qualitative comparisons against specified acceptance criteria. The evaluation criteria included speed, controls, usability, and reliability testing. Because the speed with which information travels is critical to the performance of a substation, the speed of IEC 61850 was compared to the speed of traditional digital serial communications processor controls and existing serial peer-to-peer communications.

A single-line schematic of this project is shown in Fig. 1. It includes two lines, a radial feeder, a ring bus, a transformer,

four circuit breakers, and switchgear. The following protection and monitoring devices are applied in the project’s power system:

• Line distance protection • Differential protection • Transformer protection • Bus differential protection • Radial feeder overcurrent protection • Communications processor • Computing platform

IV. IMPLEMENTATION OF IEC 61850 Using the described system, pertinent IED information was

output to an HMI created with the Wonderware® software application. The software runs on a substation-hardened rugged computing platform with a touchscreen monitor-based GUI. Using the information selected for the HMI and associated timestamps at the IED level, the team tested the speed difference between IEC 61850 communication and the typical serial connections.

There are two methods of configuring Windows® software applications to interact with the IEC 61850 protocol drivers that also run on the Windows operating system on the rugged computer. These two are DDE and OPC.

DDE offers an easy and flexible method of passing data from one Windows application to another. Given that AX-S4 MMS software (see Section IV.A for description of the software) has the capability to communicate via DDE, this method was a natural choice because Wonderware has mature and often-used DDE interfaces as well. This eliminates the need for a third-party protocol converter and reduces points of failure. In addition, Microsoft Excel® can be programmed as a DDE interface to mimic an HMI for testing controls and status through IEC 61850.

The OPC communications method was based on the OLE, COM, and DCOM technologies developed by Microsoft for the Windows operating systems.

Feeder Protection

Transformer Protection

Bus Diff Protection

Feeders

#14000 #13000 #12000 #11000

138 kV Ring Bus

Line and Diff Protection

Line and Diff Protection

1200/5

1200/5

1200/5

2000/5

Line 1Line 2

Fig. 1. One-Line Diagram

3

Fig. 2 represents the IEC 61850 process using the two different communications methods—DDE and OPC. DDE communication requires fewer individual pieces of software. The design team selected Wonderware for HMI data and control using DDE because the team was familiar with it. The user can select either method for implementation.

IED IEDIED

Switch

AX-S4 MMS(IEC 61850 Interface)

WWMME Link ReLab OPC Console

Wonderware®

DDE OPC

ReLab SLC

ClearView Server

ClearView

Fig. 2. DDE and OPC Control Implementation

A. AX-S4 MMS AX-S4 MMS is the IEC 61850 interface that acts as a

client to collect data and perform controls and as a server to provide data. It provides both DDE and OPC interfaces to the IEC 61850 protocols based on MMS. AX-S4 MMS, with the capability to communicate both DDE and OPC, provides many integration options. The software is a server that links clients such as Wonderware and Excel with MMS devices like protective relays. AX-S4 MMS provides real-time data in several ways. Data can be requested via a browser, similar to a web browser that understands and displays values and descriptions. One such browser is MMS Object Explorer (MOE). The same data, or subsets of it, are collected via MMS reports when data changes or at a fixed update rate to support an HMI like Wonderware or Excel.

Starting MOE automatically initiates AX-S4 MMS and other necessary software components to prepare the computer to act as an MMS client. The MOE is the configuration viewer of all the physical and logical devices available on the network and visible to the AX-S4 MMS server. Once MOE is configured with the device connection information, such as network IP address, all available device information from each IED can be viewed in the explorer window. This information includes the IED’s physical and logical status and analog values. Instantaneous and dead-band magnitude and angle values are available for measured analog inputs such as voltage and current. Instantaneous and dead-band values are

also available for calculated analog values such as watts, vars, and frequency. Fig. 3 illustrates the ability to see the individual frequency measurement for a line distance protection relay. This window also identifies the device name and DDE item information. The DDE item string information serves as the data name for interface with software applications.

Fig. 3. Starting AX-S4 MMS

The steps to add an IED to AX-S4 MMS are shown in Fig. 4. From the Object Explorer window, select Tools > AX-S4 MMS Configuration Utility. In the Utility window, select Configuration > Network > Addressing.

Fig. 4. Adding an IED to AX-S4 MMS

4

The Network Addressing window will appear, as shown in Fig. 5. Confirm that this window identifies Host Names and select New > Next. From the IED Identifying window you will “name” the relay. To avoid computing issues with spaces or dashes, use the underscore character (_). Then select Next. Select Yes to configure a TCP/IP network, and enter the IP Address of your host, for example, 192.168.0.90.

Fig. 5. Network Addressing

After entering the IP address, select Next; then select Add AR Name > Next. In this window, you will see the host name that you added. Select Next; enter the same name for the AR Name you are configuring and select Next > Next > Next > Finish > Finish. At this point, you may close out of the Network Addressing window and the Configuration Utility window. Bring up the AX-S4 MMS window, and select Tools > Reconfigure, as shown in Fig. 6.

Fig. 6. Reconfiguring AX-S4 MMS

This procedure will reconfigure AX-S4 MMS to accept the new changes or additions to the configuration utility. Viewing the MMS Object Explorer window, choose from the folder list on the left side of the screen, right click on Physical Devices, and select Add Nodes, as shown in Fig. 7.

Fig. 7. Adding Nodes and Updating

Adding the new IEDs through this window will update all IEDs that have been added to the remote node listing. Select OK to add the new nodes and complete this process.

From the MMS Object Explorer window, as shown in Fig. 8, if the IEDs are online with the station LAN, relay status can be polled, and AX-S4 MMS can implement controls. The next set of instructions details the steps to test the controls and status mechanisms using Microsoft Excel.

Fig. 8. Adding Polled Devices

Open a new Excel spreadsheet and start a new Visual Basic® Editor. Copy the script provided in the appendix into the Editor window and save. Then click on Button located in the Forms toolbar and assign the macro script GetFrequency. Another button can be assigned to a different macro script SendOutput1.

Using any of the Excel button makers, select a function name and cell for that button. The cell button will run that particular function outlined in the Visual Basic script.

B. Wonderware For this project, Wonderware InTouch® software was used

as an HMI to run the IEC 61850 application. Controls are administered by issuing the DDE item from AX-S4 MMS when a particular action is taken in Wonderware. Figs. 9 through 13 show how to set up a pushbutton to administer a control, such as setting and resetting a remote bit in a relay.

5

To create a new window after starting a new application in Wonderware, right click on Windows; then click on New, as shown in Fig. 9.

Fig. 9. Creating New Window in Wonderware

This action will bring up the Window Properties box, as shown in Fig. 10, where the appropriate window selections and naming convention are chosen. The user can populate the window as desired.

Fig. 10. Window Properties Screen

As seen in Fig. 11, a few simple pushbuttons and status indicators are chosen. To label the pushbuttons, click on the T in the toolbar, and place the cursor in the desired box to label it appropriately. Text can also be typed in the window to label other actions, like the LED status in the relay.

Fig. 11. Wonderware Draw Object Toolbar

The user must now configure the status indicators and pushbuttons to operate as desired. As seen in Fig. 12, select PB1, and choose Action under the Touch Pushbutton column to configure this particular rectangle as a pushbutton. In the same way, check Discrete under the Fill Color column to configure the desired status indicators.

Fig. 12. Configuring Pushbuttons

Fig. 13 shows the field in which to enter the action script.

Fig. 13. Script Activated by Pushbutton

6

The following script sets a remote bit in the relay: WWMMETopic = “AXS4MMS|Relay_2”; WWMMEItem1 = “Read AR=Relay_2 Name=RBGGIO1 Domain=Relay_2CON DTDL={(ctlVal)Bool”; WWMMEItem2 = “,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO11[Oper]] Rate=2”; SendControl = “1”; DisableButtons = 1;

This script will run a small program WWMMELink that concatenates WWMMEItem1 and WWMMEItem2 from the Wonderware script. In Wonderware, every action, status, and message has a tagname associated with it. This tagname defines what type of action will be performed as well as where it should be sent based on its “AccessName” and “Item.” The item is where the control string would be entered. However, a tagname item has an 80-character limit and the AX-S4 MMS control string is more than 90 characters. In order to overcome this obstacle, the AX-S4 MMS control string is divided into two sections in the Wonderware script. Then the WWMMELink program puts the two sections together and sends the complete control string as required to AX-S4 MMS. The WWMMELink also has the option to send a pulse, a one, or a zero.

The WWMMELink proved to be an extremely valuable tool to allow the HMI to communicate with the relays, making retrieval of status from the relay much easier. Moreover, the DDE item from AX-S4 MMS is placed into the item field for that particular tagname. The following string represents an item to retrieve frequency from a relay:

RELAY1_1MET/METMMXU1$MX$Hz$instMag This item is just one example of polling data from a relay.

All other status and indication points are retrieved in this same way. This item string can be traced in the MMS Object Explorer that is a GUI for AX-S4 MMS.

Another method of real-time data retrieval is to bundle all of the data that the HMI requires into a report within the IED. Once the client, in this case AX-S4 MMS, establishes a connection with the IED, the report will be sent any time data change. This will reduce network traffic and assure timely updates of changes. Both the polling and reporting methods are frequently used for HMI GUIs, with designers balancing their functional and performance differences.

V. TESTING The group also designed and drafted a complete set of

drawings that include ac schematics, dc schematics, one-line diagrams, communications diagrams, panel layouts, and wiring diagrams. As mentioned earlier, these drawings enabled the group to do a quantitative and qualitative compar-ison between IEC 61850 and the traditional methods of wiring microprocessor-based protective relays. The evaluation criteria were speed, controls, usability, and reliability. The speed with which information can travel is critical to the performance of a substation. The speed of IEC 61850 was

compared to traditional digital serial communications processor controls and existing serial peer-to-peer communi-cations.

One of the IEC 61850 protocols is GOOSE, which is used for substation events such as commands, alarms, and message indicators. GOOSE messaging is a key component of IEC 61850 and allows IEDs to effectively communicate to one another to accomplish interlocking and integrated protection schemes. A single GOOSE message that is sent by one IED can be received and used by several other IEDs. GOOSE takes advantage of the speed of Ethernet and supports real-time messaging, which is crucial for accurate event recording. GOOSE messages are transmitted periodically to increase the probability of delivery, as well as every time the message contents change.

A. Relay-to-Relay Communication When considering how to test the speed of GOOSE

communication, the highest accuracy is required. The project team decided to use the SER function available in the relays. The SER data are timestamped to the millisecond. Communications timing tests between the devices were completed by analyzing different substation scenarios that are commonly used in GOOSE messaging. Each test included the following:

• Hardwired I/O • Serial peer-to-peer communications protocol • IEC 61850 GOOSE messaging Fig. 14 shows the connections of the first test setup.

Relay

Relay

PORT 2

OUT02OUT01

ALARMIRIG-B

PORT 5

IN101PORT 2 PORT 5

Antenna 75'

PORT 1 PORT 2

ALARMIRIG-B

OUT101

LegendCoaxCopperEthernetSerial

Satellite Clock

Switch

Fig. 14. First Test Setup

The team assumed that the hardwired I/O would be the fastest, because there was no communications processing time in between the relays. However, as demonstrated by these and other tests performed, the physical detection circuit time takes longer than the communications message processing time to verify that an element has changed state.

To accurately test all scenarios, each component of the test must start from a single relay element to assure that they all start at the same instant in time. To initiate the test, a

7

pushbutton on the front panel of Relay 1 was programmed to set a logical latch function, PLT10, inside the relay.

PLT10SET := PUSHBUTTON10_PULSE AND NOT PLT10 (Actuating the test button sets the logic bit if it is not already set) PLT10RESET := PB10_PULSE AND PLT10 (Actuating the test button resets the logic bit if it is already set)

The latch was programmed to operate a contact output (OUT101), a logic equation to transmit a serial peer-to-peer output (TMB1A), and an IEC 61850 communications card output (CCOUT01).

OUT101 = PLT10 TMB1A = PLT10 CCOUT01 = PLT10†

The initiating latch, the contact output, the serial peer-to-peer message equation, and the IEC 61850 communications card output were added to the SER trigger list so that all state changes would be recorded.

SER1 = PLT10,PLT10,Asserted,Deasserted,N SER2 = OUT101,OUT101,Asserted,Deasserted,N SER3 = TMB1A,TMB1A,Asserted,Deasserted,N SER4 = CCOUT01,CCOUT01,Asserted,Deasserted,N

This information provides the exact time that the relay asserted each output and the latch.

The contact output and the serial peer-to-peer message output consistently assert at the same time and within the same millisecond as the initiating latch along with the IEC 61850 communications card output. This timing was consistent each time the communication was tested. In the receiving relay, elements representing the hardwired contact input, serial peer-to-peer received message, and IEC 61850 communications card input were programmed into the SER equation to record the relative time of receipt of the various signals.

SER1 = IN101,IN101,Asserted,Deasserted,N SER2 = CCIN001,CCIN001,Asserted,Deasserted,N SER3 = RMB1A,RMB1A,Asserted,Deasserted,N

To test the speed of the serial peer-to-peer communi-cations, the speed was set to its maximum of 38400 bps. The results are shown in Table I.

† CCOUT01 is used because the relay will respond faster than if not using a CCOUT equation. If the CCOUT equation is not used, the relay polls every 500 milliseconds instead of sending the message as soon as the latch changes state. This could cause a delay of up to 800 milliseconds in the communica-tions.

When sending a latched state of 1 with IEC 61850 GOOSE communi-cation using a CCOUT in the dataset, the relay sends a pulse that asserts and then deasserts the receiving relay, unlike the serial peer-to-peer and hardwired schemes where the receiving relay gets asserted and stays asserted. However, if the CCIN equation is used without the CCOUT, the receiving relay gets asserted and stays asserted, just like the serial peer-to-peer and the hardwired schemes.

TABLE I COMPARISON AT 38400 BPS

38400 Time Difference

Start LATCH10

Time

RMB1A Received

Time

IN101 Received

Time

CCIN001 Received

Time RMB1A IN101 CCIN001

25.637 25.641 25.645 25.641 0.004 0.008 0.004

54.291 54.295 54.3 54.295 0.004 0.009 0.004

47.291 47.295 47.3 47.295 0.004 0.009 0.004

4.393 4.395 4.402 4.395 0.002 0.009 0.005

38.545 38.552 38.554 38.550 0.007 0.009 0.005

23.72 23.725 23.729 23.724 0.005 0.009 0.005

54.795 54.8 54.804 54.799 0.005 0.009 0.005

22.497 22.5 22.501 22.504 0.003 0.009 0.005

54.297 54.3 54.306 54.3 0.003 0.009 0.005

29.074 29.081 29.078 29.081 0.007 0.009 0.005

8.775 8.779 8.783 8.779 0.004 0.008 0.004

43.2 43.204 43.208 43.204 0.004 0.008 0.004

11.827 11.831 11.835 11.831 0.004 0.008 0.004

39.227 39.231 39.235 39.231 0.004 0.008 0.004

10.152 10.156 10.16 10.157 0.004 0.008 0.004

39.729 39.733 39.737 39.733 0.004 0.008 0.004

10.004 10.008 10.012 10.01 0.004 0.008 0.004

38.254 38.258 38.262 38.26 0.004 0.008 0.004

12.306 12.308 12.315 12.312 0.002 0.009 0.004

2.931 2.933 2.94 2.937 0.002 0.009 0.004

Average (Seconds) 0.00400 0.00855 0.00435

At a data transfer rate of 38400 bps, serial peer-to-peer communication and GOOSE IEC 61850 protocol have about the same transmission time. Surprisingly, hardwired I/O is the slowest of the three connections.

The serial peer-to-peer connection was also tested using a channel speed of 19200 bps. With this change, the serial peer-to-peer channel speed is 2 milliseconds slower than the GOOSE messaging speed, as shown in Table II. The speed of serial peer-to-peer communication is dependent on the data transfer rate, where as IEC 61850 is not. This is another advantage of using Ethernet-based communications.

8

TABLE II COMPARISON AT 19200 BPS

19200 Time Difference

Start LATCH10

Time

RMB1A Received

Time

IN101 Received

Time

CCIN002 Received

Time RMB1A IN101 CCIN001

35.905 35.911 35.913 35.909 0.006 0.008 0.004

11.432 11.438 11.44 11.436 0.006 0.008 0.004

45.732 45.738 45.74 45.736 0.006 0.008 0.004

14.457 14.461 14.465 14.461 0.004 0.008 0.004

50.459 50.465 50.467 50.463 0.006 0.008 0.004

21.259 21.265 21.267 21.263 0.006 0.008 0.004

49.134 49.142 49.142 49.138 0.008 0.008 0.004

10.686 10.69 10.695 10.690 0.004 0.009 0.004

40.613 40.62 40.622 40.618 0.007 0.009 0.005

10.013 10.02 10.022 10.018 0.007 0.009 0.005

Average (Seconds) 0.006 0.0083 0.0042

These tests show that both serial peer-to-peer and IEC 61850 communications were faster than the hardwired connection. However, when figuring in reliability, hardwire I/O is still a good method of communication. If the Ethernet switch failed, then all protection would be lost in the IEC 61850 applications, proving that equipment reliability is crucial for any communication standard.

IEC 61850 does present some advantages when compared to the serial peer-to-peer protocol. First, the serial peer-to-peer protocol allows only eight bits to be transmitted in each direction and GOOSE messages can transmit up to 140 Boolean data elements. However, most interlocking and protection schemes require the exchange of fewer than eight bits from each IED. Second, though both can be deployed directly between two peers using a single cable, it is more useful to multicast the message to several peers with a communications switch. For each message type, a communications switch must be added with each IED connected to it via a direct link to multicast messages to all IEDs on the network. The serial switch for the peer-to-peer serial message adds latency to the message transfer on the order of 15 milliseconds at 19200 bps. Ethernet switches used for GOOSE message multicast add a very small latency, less than 0.2 milliseconds.

Fig. 15 shows the connections of a second test setup with the serial message communications switch connecting all the IEDs on the network.

Table III shows the test results with the serial message communications switch integrated into the communications scheme. As with the use of an Ethernet switch, the reliability decreases when adding another device to the system; however, the serial message communications switch is designed for the mission critical purpose of multicasting protection data and has a typical MTBF of about 300 years, contrasted with Ethernet switches that at best have an MTBF of 23 years.

Further, the peer-to-peer protocol has features to check that the channels are operational, verify their dependability, and calculate channel availability. If any channel were to fail, settings in the relay would activate an alarm. IEC 61850 has no checking features or bits to set if the channel loses communication.

When compared with direct relay-to-relay communication, multicasting through the serial message communications switch increases the delay; however, this delay is close to one electrical cycle, which is fast enough for many protection schemes.

Relay

Relay

Logic Processor

PORT 2

PORT 1 PORT 2

OUT02OUT01

ALARMIRIG-B

PORT 5

IN101PORT 2 PORT 5

Antenna 75'

Serial Message SwitchPORT 1 PORT 2

ALARMIRIG-B

OUT101

LegendCoaxCopperEthernet

Satellite Clock

Fig. 15. Second Test Setup with Serial Message Switch

TABLE III COMPARISON AT 19200 BPS WITH LOGIC PROCESSOR

19200 Time Difference

LATCH10 RMB1A RMB1A

22.872 22.887 0.015

12.699 12.714 0.015

39.576 39.591 0.015

1.326 1.343 0.017

29.151 29.168 0.017

49.001 49.016 0.015

6.353 6.368 0.015

25.878 25.893 0.015

41.553 41.566 0.013

19.228 19.241 0.013

Average (Seconds) 0.015

9

B. HMI Controls

Many substations have HMIs that allow the user to look at a computer screen and have control over the entire substation with the click of a button. HMIs usually have advanced graphics with many status indicators and pushbuttons. The user can have more control, safer operation, and faster recovery times after outages. The cost of the HMI is justified by facilitating work and analysis on the substation system.

Fig. 16 shows a setup of an HMI SCADA system with a Wonderware application to test controls of the relays using two different communications methods. One method implements IEC 61850 and the other method applies serial communication through a communications processor.

HMI

Ethernet Comm 1PC

Satellite Clock

Relay

Relay

Communications Processor

PORT 2

PORT 1 PORT 2

OUT02OUT01

ALARMIRIG-B

PORT 5

PORT 2 PORT 5

Antenna 75'

SwitchPORT 1 PORT 2

ALARMIRIG-B

LegendCoaxCopperEthernetSerial

PORT 14

PORT 3

Fig. 16. Third Test Setup

The basic setup of the two relays consisted of two different internal IED logic bits referred to as remote bits—RB01 and RB02. RB01 was sent through the Ethernet with IEC 61850. RB02 was sent through a serial connection with a communications processor as an intermediary between the HMI and as many as 16 relays. Since the times were so drastically different, the team could easily identify which was quicker by observing the LEDs on the relay. The test results were surprising again, as shown in Table IV.

TABLE IV CONTROLS FROM PC TO RELAY VIA IEC 61850 VS. SERIAL

Serial (RB02) IEC 61850 (RB01) Difference

21.853 24.538 2.685

12.755 15.517 2.762

1.748 4.506 2.758

5.962 8.494 2.532

38.554 40.473 1.919

17.575 20.433 2.858

50.893 52.483 1.59

22.777 25.389 2.612

50.795 52.568 1.773

29.449 31.372 1.923

37.342 39.34 1.998

0.715 3.332 2.617

36.088 38.311 2.223

58.978 61.309 2.331

30.59 33.242 2.652

51.088 53.173 2.085

23.519 26.221 2.702

45.596 48.211 2.615

6.446 9.177 2.731

29.265 31.152 1.887

Average (Seconds) 2.36265

On average, IEC 61850 was almost 2.5 seconds slower than communication through a serial connection with a communications processor, as configured in Fig. 16. The IEC 61850 speed is highly dependent on the script that is written to send the controls to the AX-S4 MMS server. The script was written with high consideration toward speed. However, IEC 61850 significantly lagged the serial communication due to the much larger message size and therefore processing overhead.

VI. RELIABILITY Reliability is always a major concern when adding

anything new to a power system. In these tests, IEC 61850 never missed a protection bit that was sent across the network. This is not to say there is not one single point of failure. The switch is a single point of failure, and IEC 61850 GOOSE does not have an acknowledgement mechanism to confirm to the publishing IED that the message was received. Instead, the sending relay will send its bit via multicast messages that are sent very rapidly at first, and then gradually slowing to a rhythmic heartbeat update time. The frequent publication right after the bit changes increases the likelihood that the information will get through the network to each intended user. In low-voltage distribution substations, it may be

10

feasible to only use a single, individual IEC 61850 system. However, in high-voltage transmission systems, the non-deterministic nature of IEC 61850 protocols suggest it may still be prudent to use two parallel forms of protection communications.

VII. CONCLUSION The ability to draw a comprehensive conclusion on a new

standard of communications protocol is difficult after only one project. However, IEC 61850 can be compared to serial communications within a substation as follows:

• Few software applications that support the protocols exist, and those that do are less mature than their serial counterparts.

• Fewer integration software tools exist. • The nature of multiple protocols in one standard and

the use of Ethernet offer more versatility. The controls from the HMI were a major roadblock in the

completion of the project. Initially, the design group decided to use OPC as a medium to transmit and receive data. Receiving data was quite easy; however, trying to implement controls through a protocol converter proved to be difficult at best, and in the end unattainable for this team. This issue was overcome by the following:

• Communicating with DDE directly to AX-S4 MMS. • Creating an HMI with Excel software. • Learning Visual Basic script. • Eliminating Wonderware limitations with Visual

Basic programming. Although the bulk of the code could be copied into each

HMI script, the code used to transmit data packets over Ethernet was unique for each relay and each type of control issued.

The performance of IEC 61850 was not an issue. The ability for companies to implement systems communicating IEC 61850 protocols with less engineering effort than previous methods will be most crucial to its success. Apply the following to decrease implementation time:

• Software to automate integration. • Tools to read configuration files and format data to

use the IEC 61850 nomenclature and constructs. • Clients and servers, such as AX-S4 MMS, tightly

integrated into HMIs and other software applications. Controlling a system through an HMI with IEC 61850 can

be as easy as integrating a new relay into the system. However, a simple and cost-effective option is to forgo the HMI, implement controls through the relay pushbuttons, and use only the IEC 61850 GOOSE protocol for peer-to-peer communication. In addition, the speed of GOOSE messaging proved to be as quick and reliable as any protocol available today. The ability to utilize TCP/IP protocol is the next logical step for future integration projects, providing quick and easy updates through remote terminal units for engineers to receive updates over any LAN.

VIII. APPENDIX The script provided here retrieves the status and sends

control to the relays using IEC 61850 through DDE communi-cation.

‘This script is created to assert RB10 to ‘1’ and then deassert RB10 to ‘0’ using Excel. Sub SendOutput1() ‘starting the function

channel_5 = DDEInitiate(“AXS4MMS”, “Relay_1”) ‘setting a channel (any name) to initiate a DDE

communication from Excel to AXS4MMS Relay_1 Application.Worksheets(“Sheet1”).Activate ‘Activating Sheet1 of the Excel application result = DDERequest(channel_5, “Read AR=Relay_1

Name=RBGGIO1 Domain=Relay_1CON DTDL={(ctlVal)Bool,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO10[Oper]] Rate=2”)

‘Setting an array called ‘result’(any name) to the requested data from AXS4MMS through the established communications channel using “Read + the DDE item string corresponding to the desired Oper”

Range(“A20:A26”) = result ‘Setting ‘result’ which contains the values of (ctlVal,

orCat, orIdent, ctlNum, T, Test, Check) to Cells in Excel (for example from A20 to A26)

[A20] = 1 ‘Setting the value of cell A20 to 1 corresponding to ctlVal

[A25] = 0 ‘Setting the value of cell A25 to 0 corresponding to Test because it needs to be a (1 or 0) instead of (True or False)

DDEPoke channel_5, “Write AR=Relay_1 Name=RBGGIO1 Domain=Relay_1CON DTDL={(ctlVal)Bool,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO10[Oper]] Rate=2”, Range(“A20:A26”)

‘Sending a poke command (write) to AXS4MMS through the set channel, using “Write + the DDE item string corresponding to the desired Oper”. The values sent are the ones stored in the corresponding Excel cells A20 to A26. Basically setting RB10 to ‘1’

[A20] = 0 ‘Resetting the value of cell A20 to 0 corresponding to ctlVal (this step is optional, it is just to reset RB10 to ‘0’)

DDEPoke channel_5, “Write AR=Relay_1 Name=RBGGIO1 Domain=Relay_1CON DTDL={(ctlVal)Bool,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO10[Oper]] Rate=2”, Range(“A20:A26”)

‘Sending a poke command (write) to AXS4MMS through the set channel, using the same DDE item string to reset RB10 to ‘0’

DDETerminate channel_5 ‘Terminating the communications channel.

End Sub

11

‘This script is created to assert RB11 to ‘1’ and then deassert RB11 to ‘0’ using Excel

Sub SendOutput0() channel_6 = DDEInitiate(“AXS4MMS”, “Relay_1”) Application.Worksheets(“Sheet1”).Activate result = DDERequest(channel_6, “Read AR=Relay_1

Name=RBGGIO1 Domain=Relay_1CON DTDL={(ctlVal)Bool,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO11[Oper]] Rate=2”)

Range(“A27:A33”) = result [A27] = 1 [A32] = 0

DDEPoke channel_6, “Write AR=Relay_1 Name=RBGGIO1 Domain=Relay_1CON

DTDL={(ctlVal)Bool,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO11[Oper]] Rate=2”, Range(“A27:A33”)

[A27] = 0 DDEPoke channel_6, “Write AR=Relay_1

Name=RBGGIO1 Domain=Relay_1CON DTDL={(ctlVal)Bool,(origin){(orCat)Byte,(orIdent)OVstring64},(ctlNum)Ubyte,(T)Utctime,(Test)Bool,(Check)BVstring2} ADL=CO[SPCSO11[Oper]] Rate=2”, Range(“A27:A33”)

DDETerminate channel_6 End Sub ‘This script is created to bring the value of frequency

from AXS4MMS using Excel Sub GetFrequency() ‘starting the function channel_3 = DDEInitiate(“AXS4MMS”, “Relay_1”) ‘Setting a channel (any name) to initiate a DDE

communication from Excel to AXS4MMS Relay_1 Application.Worksheets(“Sheet1”).Activate ‘Activating Sheet1 of the Excel application fq = DDERequest(channel_3, “Read AR=Relay_1

Name=METMMXU1 Domain=Relay_1MET DTDL=Float ADL=MX[Hz[instMag[f]]] Rate=2”)

‘Setting a variable called ‘fq’(any name) to the requested data from AXS4MMS through the established communications channel using “Read + the DDE item string corresponding to the desired Oper”

Sheet1.Cells(14, 1) = fq ‘Setting ‘fq’ which contains the value of the frequency

to Cell in Excel (for example A14) DDETerminate channel_3 ‘Terminating the

communications channel End Sub’

IX. BIOGRAPHIES Youssef Botza earned his BS in Electrical Engineering from the University of North Carolina at Charlotte with honors, where he served as a treasurer for the IEEE UNC Charlotte Chapter in 2006. Youssef joined Schweitzer Engineering Laboratories, Inc. as an intern in 2006. In 2007, he was promoted to an associate protection engineer in the systems and services division. His responsibilities include design, development, testing, and commissioning of substation control house systems.

Matthew Shaw holds an AS degree from Catawba Valley Community College. Currently he attends the University of North Carolina Charlotte, working toward a BS degree in electrical engineering. Matthew is an active member of Tau Beta Pi Engineering Honor Society and serves as vice president of the IEEE Student Chapter. Presently Matthew is working at Schweitzer Engineering Laboratories, Inc. as an engineering intern, gaining valuable knowledge in power systems and substation integration.

Peter Allen earned his BS in Electrical Engineering from the University of North Carolina at Charlotte. In 2006, he was employed by Central Piedmont Community College as an instructor in their Engineering Technology department. In 2007, he joined Schweitzer Engineering Laboratories, Inc. as an intern and was later hired as an associate automation engineer. Peter has been a member of IEEE since 2005.

William Michael Staunton expects to complete his BS in Electrical Engineering degree at the University of North Carolina Charlotte in the summer of 2008. Since 2006, Michael has been an active member of the North Carolina Society of Engineers. In 2007, he worked as an electrical design intern at Automation Tooling Systems Carolina, Inc. Currently, Michael is an intern at Schweitzer Engineering Laboratories, Inc. in the Charlotte office, working with protection engineers to design and complete substation protection solutions.

Dr. Robert Cox received his BS, MEng, and PhD degrees from Massachu-setts Institute of Technology in 2001, 2002, and 2006. He is currently an assistant professor of electrical and computer engineering at UNC Charlotte. His research is focused on the design, analysis, and maintenance of electrical actuators, power-electronic drives, analog instrumentation, and sensors.

Michael Boughman earned his BS in Electrical Engineering from the University of North Carolina at Charlotte. In 1987, he was employed by Duke Energy Company, where he worked as a technical specialist. At Duke Energy, his responsibilities included substation integration/automation design and implementation. In 1999, he joined Schweitzer Engineering Laboratories, Inc. (SEL) as an integration application engineer. His responsibilities include technical support, application assistance, and training for SEL customers.

Casey Roberts earned his BS in Electrical Engineering from the University of North Carolina at Charlotte. In 2004, he was employed by Duke Power as an intern in the Fossil/Hydro Generation division. At Duke, his responsibilities included development of controls and HMI for the overall plant distributed control system. In 2006, he joined Schweitzer Engineering Laboratories, Inc. as an automation intern. As of 2007, he was promoted to an associate automation engineer. His responsibilities include design, development, testing, and commissioning of SCADA and HMI systems. Casey has been a member of IEEE since 2007.

William Rominger earned his BS in Electrical Engineering from the North Carolina State University. In 2003 he was employed by ABB as an intern in the Protection and Control department. At ABB his responsibilities included ac and dc schematic design and implementation. In 2006, he joined Schweitzer Engineering Laboratories, Inc. as an associate protection engineer. His responsibilities include design, settings calculations and implementation, testing, commissioning, and customer training. William has been an IEEE member since 2006.

© 2008 by University of North Carolina Charlotte and Schweitzer Engineering Laboratories, Inc.

All rights reserved. 20080417 • TP6309-01